Stock Market Frenzy, or How to Make Onions of Tulips

“Blaine’s Travels” informs us that in 1634 a ship’s captain made a delivery of the Levant’s richest silks to a Dutch merchant in the city of Haarlem. The delighted merchant entertained the captain at his place of business and made him a gift of a fine red herring for his breakfast. The captain, seeing what appeared to be an onion lying oddly out of place amid the merchant’s wares, slipped it into his pocket to eat with his herring.

The next morning, as the captain sat on his bridge, contentedly downing the last of the herring and onion, the outraged merchant arrived with the local gendarmerie to clap the poor man in prison. The “onion” was in fact a tulip bulb, worth at the moment some 4,000 florins, or enough, the merchant wailed, to “have sumptuously feasted the Prince of Orange and the whole court of the Stadtholder.”

No one has ever explained the frenzy for tulip bulbs that swept through Holland and most of the rest of Northern Europe in the mid-17th Century. The enterprising Dutch, however, did not trouble over explanations, but plunged instead into the business of buying and selling bulbs. The Dutch tulip exchange at one point rivaled the London Stock Exchange in volume and sophistication. Sharp-eyed analysts made tidy fortunes by locking in minuscule spreads as they bought and sold huge volumes of bulb contracts for future delivery or sale.

The intelligence, insight and rigorous trading skills needed to succeed on the Dutch tulip exchange bore no relation to the intrinsic value of the bulbs. To a sharp trader, intrinsic value is irrelevant; it is movement that counts. That is as true today on Wall Street, where a new generation of traders who took their breakfast Pablum topped with bits and bytes and whose formative years were spent peering at “Space Invader” video screens, are revolutionizing the financial markets.

Fundamental changes in world financial markets, combined with the remarkable power of the high-speed computer, have introduced enormous volatility into stock and bond prices, as “program trades.” Such trades are directed by a computer programmed to spot tiny price anomalies, to cause buying and selling huge blocks of securities within seconds, making prices careen up and down as the old-timers run for their Maalox.

Over the last decade or so, federal pension legislation has required the accumulation of enormous securities holdings in corporation pension funds, while the abolition of fixed commissions has made it very cheap to buy and sell huge blocks of stock. At the same time, the poor track record of the typical portfolio manager in guessing the direction of prices has caused a search for techniques that will minimize the possibility of portfolio losses.

The search for safety has spawned a plethora of new financial instruments--stock index shares, index options, index futures--and a variety of computerized “portfolio insurance” strategies, all designed to minimize risk and to create opportunities for a profit no matter which way the market moves.

To begin with, big investors in both stocks and bonds are increasingly hedging their risk by buying indexes rather than individual securities. Instead of guessing how individual issues will perform, their computers apply statistical techniques to pick a mix of securities that matches a popular index, like the Standard & Poor’s 500 Composite Index or the Shearson-Lehman Government/Corporate Bond Index. The investor has thereby limited his risk to the overall trend in the market and is no longer exposed to random events in an individual company.

The next step is to hedge the portfolio with futures. If an investors owns $10 million worth of stock index shares at $100 apiece, he can lock in that value for a month by selling $10 million worth of stock index futures at 100 for delivery 30 days later. He has made a contract, in effect, to sell the same number of shares as he owns at his present price, so he is protected against a loss--and has sacrificed the opportunity for a big gain--at the cost of a small transaction fee.

“Portfolio insurance” computerizes the hedging strategy. If an index portfolio slides in value past a trigger point, usually 3%-4%, the computer starts selling future contracts to protect against a further loss. As the portfolio starts to edge back up in value, the computer will start to buy futures contracts to “unwind” its position. If it’s contracted to sell in the future at 100 on the way down, for instance, it will contract to buy at 100 on the way back up. The two futures positions will then offset each other and, again except for the transaction fees, will not act as a ceiling on the investor’s gains if his portfolio keeps on rising.

Life gets interesting when the value of index futures and index shares slip out of line with each other, which often happens. Futures trade primarily on the Chicago Mercantile Exchange, where trading habits are looser, freer and faster than on the New York Stock Exchange, and where prices react more quickly to rumors or international developments. To the cold, diode-eyed computers, each little price anomaly presents an opportunity to plunge after a quick profit.

If a program trader holds stock index shares at 100, for example, and index futures creep up to 101, his computer will instantly start to sell futures, locking in a 1% gain on the portfolio. Assume the shares then start to fall, to 98, say, but that the futures, as often happens, fall slightly faster, to 97. The computer will switch strategies and sell the shares to buy the cheap futures contracts. The total profit will then be 2%: the difference between the futures sale contract at 101 and the buy contract at 97, offset by the loss on the share price drop from 100 to 98.

The small spreads available in index arbitrage mean that the traders must operate in huge blocks, or else the exercise will not be worth the effort. And that is why computerized trading can take on the life of its own. Once the market starts down, program traders can give it an extra push in their anxiety to sell futures contracts. The rush to sell will make futures contracts cheap, so stock index funds will start selling huge blocks of shares to buy the cheaper futures. The stock sales drive prices down even faster, so the program traders’ computers start selling futures more frantically than ever, driving prices down that much more--and so on. Something of that sort happened on the day the Dow took its remarkable 86-point plunge earlier this month.

Four times a year, index futures, index options and stock options expire at the same time, so the spreads between these three instruments and the underlying stocks narrow to zero. This is the “triple witching hour.” When traders frantically try to unwind their positions in the last minutes, huge volumes of trading can be generated. The Sept. 19 “witching hour,” however, was so widely anticipated it became a non-event.

The brokerage houses’ own computerized trading systems can cause them huge losses. When the computers all start flashing sell signals, the brokers have no choice but to buy falling stock for their own accounts. All of their big customers will be selling at the same time, so there will be no offsetting buyers. The brokers must either buy and watch their capital erode or risk alienating their most important clients.

In theory, computerized trading should eliminate volatility by bridging price discrepancies. In practice, it doesn’t seem to work that way. Perhaps the dissociation of trading strategies from underlying portfolio values may be part of the problem. Some of the most brilliant traders in the 17th Century, for instance, woke up eventually to discover that their cleverest strategies produced, in the final analysis, only a handful of onions.